The association between systolic blood pressure reduction during clonidine suppression testing and the decrease in plasma catecholamines and metanephrines

Abstract

Borderline isolated norepinephrine (NE) and normetanephrine (NMT) elevation is common among patients with suspected pheochromocytoma and paraganglioma (PPGL). The clonidine suppression test (CST) may help establish the etiology in these cases. Prolonged laboratory processing and/or paucity of reliable biochemical assays may limit the utility of CST. The aim of this study was to evaluate whether blood pressure (BP) reduction during CST is associated with alterations in plasma NMT/NE, thereby potentially providing an immediate indication of CST results. In this cross‐sectional study, the authors included all consecutive patients with suspected PPGL who underwent CST from January 1, 2014, to December 31, 2019. Linear regression models were conducted to evaluate the association between BP reduction and decrease in plasma NMT/NE. The final analysis included 36 patients (17 males). The decrease in systolic BP (SBP) 90 minutes postclonidine was associated with a decrease in plasma NMT (R = 0.668, P = .025) and NE (R = 0.562, P = .005). A 40% decrease in NMT and NE correlated with a 9.74% and 7.16% decrease in SBP, respectively. Subgroup analyses demonstrated that the association between SBP reduction and the decrease in plasma NMT (R = 0.764, P = .046) and NE (R = 0.714, P = .003) strengthens among patients with hypertension and among those with diabetes mellitus (R = 0.974, P = .026 for NMT). In conclusion, SBP reduction during CST is associated with plasma NMT and NE decrease. Therefore, the decrease in SBP 90 minutes postclonidine may serve as an immediate complementary clinical tool for PPGL diagnosis.

1. INTRODUCTION

Pheochromocytoma and paragangliomas (PPGLs) are catecholamine producing neoplasms, originating from the chromaffin cells of the adrenal medulla or from extra‐adrenal sympathetic ganglia, respectively. ¹ PPGL prevalence in the hypertensive population is estimated between 0.2% and 0.6%. ² If left untreated, PPGLs are associated with an increased risk for cardiovascular morbidity and mortality; thus, early diagnosis and treatment are imperative. ³ , ⁴

The recommended screening tests for PPGLs include measurement of plasma or urine metanephrine levels (metanephrines and normetanephrines [NMT], collectively referred to as Mets). ⁵ , ⁶ 25% of patients screened for PPGLs have borderline biochemical results, which are inconsistently defined as elevated Mets up to 2‐3 times the upper normal limit. ⁷ False borderline elevation of Mets may be due to increased sympathetic activity (ie, uncontrolled essential hypertension [HTN], emotional stress, certain medications, and specific dietary intake) and/or poor test performing technique. ⁸ , ⁹ , ¹⁰ In these cases, the clonidine suppression test (CST) may assist in distinguishing between true‐ and false‐positive results with high sensitivity and specificity, particularly when isolated NMT elevation is observed. ⁹ , ¹¹ , ¹²

The CST has several limitations. First, in many low‐volume laboratories, there could be a potential delay in receiving NMT results due to the required laboratory processing time. In addition, older biochemical assays for analyzing plasma Mets, which are still utilized by many centers, lack the specificity of LC‐MS/MS assays. ¹² , ¹³ Therefore, more readily available surrogate clinical markers for NMT suppression in CST may be beneficial. Previous studies have demonstrated inconclusive findings regarding the possible association between BP alterations postclonidine administration and the biochemical results of CST. ¹⁴ , ¹⁵ Moreover, these studies are relatively old and as such measured only catecholamine levels, that is, norepinephrine (NE) levels, rather than the more precise NMT measurement. ⁵

Therefore, in this study the authors aim to evaluate whether BP alterations during CST are associated with a decrease in plasma NMT and NE levels.

2. METHODS

2.1. Study population

This is a retrospective cross‐sectional study conducted at the Division of Endocrinology, Diabetes and Metabolism, at the Chaim Sheba Medical Center, Israel. Our study population included all consecutive patients who underwent CST between January 1, 2014, and December 31, 2019. The Helsinki committee approved this study with no requirement for participants' informed consent, given the assurance of strict maintenance of participants' anonymity during data extraction and analyses.

Overall, 49 patients' files were evaluated, of whom 36 were included in the study. Patients were referred for CST by their physicians due to elevated plasma and/or urine catecholamine and/or metanephrine levels. We excluded subjects for whom there was no complete record of biochemical test results, blood pressure (BP) measurement, or those whose CST was not performed as recommended by the 2014 Endocrine Society (ES) guidelines ¹² (Figure 1). Briefly, a venous cannula was placed in an antecubital vein. After 20 minutes of supine rest, a first blood sample is drawn. Clonidine was administered orally at a dose of 0.3 mg. Blood pressure and heart rate were measured in the supine position, using the “Welch‐Allyn” Vital Signs Monitor (Tiger‐Medical), at 30‐minutes intervals before and during the test. All BP measurements were obtained as single measurements. Three hours after drug administration, a second blood sample is drawn. The tubes with blood samples are immediately placed on ice, and blood samples are analyzed for plasma normetanephrine.

FIGURE 1.

Flowchart of the patients included in the study

2.2. Study variables

Age was analyzed as a continuous variable. Body mass index (BMI) was calculated based on measured height and weight and expressed in units of kg/m^2.

Biochemical test results were expressed as the percent decrease of plasma NMT or NE levels. In all CST conducted prior to May 2018, NE levels were utilized. In CST conducted after May 2018, NMT levels were primarily used. The percent decrease was calculated as the percent change in NMT or NE level between baseline levels and 3 hours postclonidine administration ($\frac{1\text{st meas}.‐2\text{nd meas}.}{1\text{st meas}.}\times 100\%$).

Systolic BP (SBP), diastolic BP (DBP), and HR were measured every 30 minutes. A percent decrease in SBP and DBP was calculated as the percent change between baseline BP and the BP every 30 minutes thereafter until CST was completed.

2.3. Laboratory techniques

Plasma NE levels were assayed by batch alumina extraction followed by liquid chromatography with series electrochemical detection. ¹⁶

Free plasma NMT was assayed using a Chromsystems reagent kit (catalogue no.cs‐81000), Multilevel plasma Calibrator set (catalogue no.cs‐81039), and free plasma NMT control levels I, II, and III (catalogue no. cs‐0384, 0385, and 0386). Quantitation of analytes was performed using UPLC‐MS/MS apparatus. ¹³ , ¹⁷

2.4. Statistical analysis

The SPSS software, version 23 (IBM), was used for all statistical analyses. We conducted linear regression models to evaluate the association between NE and/or NMT percent decrease and the percent decrease in SBP and DBP, as continuous variables. The percent decrease in SBP and DBP was evaluated separately for every time point in 30‐min intervals. Subgroup analyses were conducted, including only patients with history of HTN, patients that were treated with at least 3 groups of BP‐lowering medications, and patients with type 2 diabetes mellitus (T2DM).

3. RESULTS

3.1. Baseline characteristics

A total of 36 patients were included in the study; 17 (47.2%) were males (Table 1). The mean age was 59 ± 12.2 years and mean BMI was 31.2 ± 5.1 kg/m² (Table 1). The indications for the CST included adrenal incidentalomas (n = 9, 25.0%), symptoms of PPGLs (n = 10, 27.8%), or were not specified (n = 17, 47.2%). Twenty‐two (61.1%) patients had HTN, 13 (36.1%) had hyperlipidemia, and 7 (19.4%) had a previous diagnosis of T2DM (Table 1). Additional baseline demographic and clinical characteristic are presented in Table 1(Supinfo S1).

TABLE 1.

Patients' baseline characteristics

	Count	N %	Mean (SD)
Sex
Male	17	47.2%
Age at time of test			59.0 (12.2)
Medical history
HTN	22	61.1%
Hyperlipidemia	13	36.1%
T2DM	7	19.4%
OSA	5	13.9%
IHD	5	13.9%
CVA	1	2.8%
CHF	1	2.8%
BP‐lowering medications
Total	24	66.7%
Beta‐blocker	16	44.4%
ACEi/ARB	13	36.1%
CCB	12	33.3%
Alpha blocker	8	22.2%
MRA	5	13.9%
HCTZ	4	11.1%
Smoking	7	19.4%
BMI (kg/m2)			31.2 (5.1)
Indication for biochemical evaluation
Adrenal incidentaloma	9	25.0%
Symptoms of PPGL a	10	27.8%
Other	17	47.2%
Systolic BP mm Hg			139.8 (16.8)
Diastolic BP mm Hg			79.1 (9.5)
Baseline measurements
HR			76 (13)
NE			1.10 b (0.5)
NMT			1.55 b (1.2)

Abbreviations: ACEi, angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; BMI, body mass index; BP, blood pressure; CCB, calcium channel blockers; CHF, congestive heart failure; CST, clonidine suppression test; CVA, cerebrovascular accident; HCTZ, hydrochlorothiazide; HR, heart rate; HTN, hypertension; IHD, ischemic heart disease; MRA, mineralocorticoid receptor antagonist; NE, norepinephrine; NMT, normetanephrine; OSA, obstructive sleep apnea; PPGL, pheochromocytoma and paraganglioma; SD, standard deviation; T2DM, type 2 diabetes mellitus.

^aSymptoms were defined as one or more of the following: paroxysmal HTN, sweating, headaches, tachycardia, palpitation, trembling.

^bLevels are expressed as times above the upper limit of normal.

3.2. Clonidine suppression test results

Twenty‐six (72.2%) CST were conducted using NE levels and 11 (28.8%) using NMT levels (one patient underwent two tests—one using NE, and another using NMT).

The mean percent decrease in NE and NMT was 53.4% ± 18.48 and 61.8% ± 27.8, respectively. Overall, all of our study participants fulfilled the pre‐specified criteria for NMT suppression postclonidine exposure and none were diagnosed with PPGL.

The mean percent decrease in SBP and DBP, for every time point, is presented in Table 2. The mean percent BP decrease from baseline was maximal 120 minutes postclonidine, for both SBP and DBP (22.1% ± 12.8 and 21.4% ± 14.3, respectively). The mean SBP and DBP over time are presented in Figure 2.

TABLE 2.

Percent blood pressure decrease during the clonidine suppression test

Time (minutes)	N	SBP % decrease		DBP % decrease
		Mean	SD	Mean	SD
60	36	16.5	12.4	15.3	13.4
90	33	20	11.2	17.8	13.6
120	35	22.1	12.8	21.4	14.3
150	35	21.7	10.8	19.8	11.3
180	34	20.6	10.2	19.2	10.9

Abbreviations: DBP, diastolic blood pressure; SBP, systolic blood pressure.

FIGURE 2.

Mean (A) systolic and (B) diastolic blood pressure trends over time (clonidine administered at time 0). Error bars ‐ ±2 standard deviation. Abbreviations: SD, standard deviation; Sys BP, systolic blood pressure; Dia BP, diastolic blood pressure

A linear regression model showed that the strongest correlation between percent decrease in SBP and the decrease in NMT and NE was 90 minutes postclonidine (Table 3). The correlation between percent decrease in SBP 90 min postclonidine with percent decrease in NMT and NE was R = 0.668 (P = .025) and R = 0.562 (P = .005), respectively. Figure 3 displays a positive linear relationship between NMT and NE percent decrease as the dependent variable, and the percent SBP decrease 90 minutes postclonidine as the independent (predictor) variable. The regression equations were determined as: y = 15.85 + 2.48x (R = 0.668) for NMT as the dependent variable (Figure 3A) and y = 35.2 + 0.67x (R = 0.562) for NE (Figure 3B). Based on these equations, a 40% decrease in NMT and NE levels correlated with a 9.74% and 7.16% decrease in SBP 90 minutes postclonidine, respectively.

TABLE 3.

Correlation between norepinephrine and normetanephrine percent decrease and blood pressure percent decrease at different time points

Time (minutes)		SBP % decrease		DBP % decrease
		% decrease NE	% decrease NMT	% decrease NE	% decrease NMT
60	R	0.290	0.479	0.098	0.241
	P	0.151	0.136	0.633	0.476
90	R	0.562**	0.668*	0.373	0.241
	P	0.005	0.025	0.080	0.475
120	R	0.378	0.253	0.078	0.055
	P	0.063	0.453	0.713	0.873
150	R	0.455*	0.420	0.153	0.505
	P	0.022	0.198	0.466	0.113
180	R	0.005	−0.179	−0.167	−0.099
	P	0.982	0.621	0.425	0.785

Abbreviations: DBP, diastolic blood pressure; NE, norepinephrine; NMT, normetanephrine; P, p value; R, correlation coefficient; SBP, systolic blood pressure.

^*Correlation is significant at the 0.05 level (2‐tailed).

^**Correlation is significant at the 0.01 level (2‐tailed).

FIGURE 3.

Linear regression plot between systolic blood pressure percent decrease at 90 min, and (A) NMT percent decrease R = 0.668, P = .025, (B) NE percent decrease, R = 0.562, P = .005. The reference line marks 40% decrease, the threshold above which CST is considered positive. Abbreviations: NE, norepinephrine; NMT, normetanephrine

3.3. Subgroup analyses

A subgroup analysis including patients with history of HTN showed a significant association between SBP percent decrease 90 min postclonidine and percent decrease in plasma NMT (R = 0.763, P = .046) and plasma NE (R = 0.714, P = .003) (Table 4). The association remained significant for percent decrease in plasma NE when analyzing only patients treated with at least 3 BP‐lowering medications (R = 0.816, P = .0013) (Table 4).

TABLE 4.

Correlation between norepinephrine and normetanephrine decrease and blood pressure reduction among hypertensive and diabetic patients

	% decrease NE	% decrease NMT
Total study population
R	0.562**	0.668*
P	0.005	0.025
Patients with HTN
R	0.714	0.764*
P	0.003	0.046
Patients with T2DM
R	0.539	0.974*
P	0.461	0.026

Abbreviations: HTN, hypertension; NE, norepinephrine; NMT, normetanephrine; P, p value; R, correlation coefficient; T2DM, type 2 diabetes mellitus.

^*Correlation is significant at the 0.05 level (2‐tailed).

^**Correlation is significant at the 0.01 level (2‐tailed).

A subgroup analysis including patients with history of T2DM showed a significant association between SBP percent decrease 90 min postclonidine and percent decrease in plasma NMT (R = 0.974, P = .026) (Table 4).

4. DISCUSSION

In this study, we demonstrate a significant positive moderate correlation between percent decrease in SBP postclonidine administration and the percent decrease in both NMT levels and, to a lesser extent, in NE levels. We found that the strongest correlation was observed 90 minutes after clonidine administration. In subgroup analyses, we observed stronger correlations between SBP decrease and decrease in plasma NE/NMT for patients with prior history of HTN and T2DM.

Mets are inactive metabolites of catecholamines. However, while catecholamines secretion is pulsatile, Mets are synthesized and secreted continuously. ⁵ Consequently, compared to catecholamine levels, Mets levels have superior sensitivity and specificity for PPGL detection. ¹⁸ , ¹⁹ According to the most recent ES guidelines, a decrease of at least 40% in NMT or catecholamine levels postclonidine administration is highly supportive of a non‐PPGL‐related etiology for NMT elevation. ¹² In a linear regression model, we calculated the SBP percent decrease 90 minutes postclonidine correlating with a 40% decrease in NMT and NE levels to be 9.74% and 7.16%, respectively. Nevertheless, despite the high sensitivity and specificity of the CST, ⁹ , ¹² there are data supporting higher false‐positive and false‐negative results, when NE levels, instead of NMT, are utilized. ²⁰ , ²¹ , ²² These reports concur with our findings regarding the superior correlation rate between SBP decrease and the decrease in NMT levels compared to NE levels. Precise and standardized biochemical assays for NMT levels (ie, LC‐MS/MS‐based assays) are not always readily available, especially in relatively small low‐volume laboratories, and even when available, results are often delayed due to technical issues. Thus, in these cases, SBP percent decrease 90 minutes postclonidine administration may potentially serve as an adjunct surrogate clinical tool in unraveling the source of borderline catecholamine elevations.

A subgroup analysis including only patients with HTN yielded a stronger correlation between mean SBP decrease and the decrease in both plasma NMT and NE. The correlation with NE percent decrease was found to be even stronger when including only patients treated with a combination of at least 3 BP‐lowering medications. Although the pathogenesis of essential HTN is not yet fully comprehended, sympathetic hyperactivity is regarded as an important contributor to the disease etiology and progression. ²³ , ²⁴ There is evidence for higher levels of metanephrines and catecholamines in patients with essential HTN, ²⁵ frequently represented as a laboratory test result above the normal upper limit. ²⁶ This could be explained by few possible mechanisms, including increased sympathetic outflow and altered synthesis and metabolism of NE. ²⁷ Furthermore, a higher central sympathetic activity was demonstrated in complicated and essential HTN, in comparison with white coat HTN or high‐normal pressure. ²⁸ These studies correspond with our finding of the stronger correlation between mean SBP decrease and the decrease in both plasma NMT and NE, found in patients with HTN, and even stronger correlation in patients treated with at least 3 BP‐lowering medications, possibly suggesting a more complicated HTN. Therefore, sympathetic‐associated catecholamine secretion is potentially more amenable to clonidine suppression in patients with resistant or complicated HTN, consequently yielding stronger correlation rates between SBP decrease and NE/NMT levels.

In an additional subgroup analysis including only patients with T2DM, we demonstrated a significantly stronger correlation between SBP percent decrease and a percent decrease in NMT levels in patients with T2DM compared to those without T2DM. Notably, all patients with T2DM had a concurrent diagnosis of HTN. Several studies have previously reported a high incidence of T2DM in patients harboring a PPGL, largely attributed to a catecholamine‐induced insulin resistance or insulin suppression. ²⁹ , ³⁰ Similarly, there is substantial evidence to support insulin‐induced sympathetic overactivation, ³¹ , ³² which likely partially explains the well‐established association between insulin resistance and HTN. ³³ , ³⁴ The stronger association between SBP decrease and a decrease in NE/NMT levels observed among patients with T2DM in our study may be explained by insulin‐induced sympathetic overstimulation resulting in higher sympathetic activity. ³⁵ As with the HTN subgroup analysis, our findings of a stronger correlation between SBP decrease and NE/NMT levels may imply that the sympathetic‐associated catecholamine secretion is more susceptible to clonidine suppression in patients with HTN and T2DM.

Our study has several limitations. First, this is an observational retrospective analysis which includes a modest number of participants. However, compared to previous similar studies ³⁶ , ³⁷ , ³⁸ , ³⁹ our cohort is relatively large. In addition, our cohort comprised participants of single institution; thus, it is unclear whether our findings can be generalized to other populations. Secondly, preclonidine levels of NE and/or NMT in many of our patients were found within normal limits. Thus, in retrospect, the CST was not indicated in these patients. However, these patients were referred to our tertiary medical center for further evaluation of elevated urine and/or plasma NMT/NE conducted elsewhere. One possible explanation to account for this discrepancy is that, contrarily to most laboratories in Israel, we use LC‐MS/MS‐based assays to analyze NMT levels. These assays have been previously shown to have superior sensitivity and specificity compared to other analytical methods. ¹² , ¹³ Thirdly, none of our patient failed to suppress NE and NMT levels postclonidine and none were consequently diagnosed with PPGL. While it would be intriguing to sub‐analyze patients based on the existence of PPGL, we aimed to establish the correlation between BP alterations and NE/NMT decrease during CST regardless of the etiology of NE/NMT elevation. We hypothesize that in cases where there is insufficient NE/NMT decrease (ie, in PPGL), the SBP response would follow a similar pattern. Although the literature on the SBP response during CST is scarce, our assumption concurs with data from a previous study by McHenry, et. al. ¹⁵ The authors have shown that the mean decrease in SBP 2 hours postclonidine administration was 10% and 21% for patients with PPGL compared to those without PPGL, respectively. Notably, NE levels postclonidine followed a similar pattern to SBP reduction with respect to PPGL status. ¹⁵ Fourthly, most of our patients (n = 24) were treated with antihypertensive medications and specifically with sympathetic antagonists (n = 19). While ES guidelines require the discontinuation of sympathetic modulators prior to the CST, ¹² we cannot exclude non‐adherence to the given instructions. Finally, the majority of the study participants were using antihypertensive medications which could potentially negate the independent BP‐lowering effects effect of clonidine. Notably, current ES guidelines permit the use of most BP‐lowering drug classes during CST. ¹² Moreover, even though most of our patients were treated with antihypertensive medications, we observed a significant correlation between the decrease in SBP and the decrease in NMT and NE levels. These findings suggest that SBP changes during CST may be informative even while patients use antihypertensive drugs.

In conclusion, we show a significant moderate correlation between SBP percent decrease postclonidine and percent decrease in the plasma NMT and to a lesser extent in NE levels. The correlation strengthens among patients with HTN, especially those treated with several BP‐lowering drugs and among patients with T2DM. Therefore, in analyzing CST results, the percent decrease in SBP 90 minutes postclonidine administration may serve as a complementary and relatively immediate tool, especially in cases where precise NMT and NE analysis assays are scarce or in time‐sensitive cases when biochemical results are delayed. Further large well‐designed and powered prospective trials are needed in order to establish the clinical utility of our findings.

CONFLICTS OF INTEREST

All authors of the manuscript “The Association between Systolic Blood Pressure Reduction during Clonidine Suppression Testing and the Decrease in Plasma Catecholamines and Metanephrines" have taken care to ensure the integrity of their work and their scientific reputation. There are no financial or other relationships that might lead to a conflict of interest.

AUTHOR CONTRIBUTIONS

All authors contributed to the study conception and design. Material preparation, data collection, and analysis were performed by TG, BF, GS, EG. The first draft of the manuscript was written by TG, BF GS, YS, and AL, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Supporting information

Supinfo S1

Golani T, Fishman B, Sharabi Y, et al. The association between systolic blood pressure reduction during clonidine suppression testing and the decrease in plasma catecholamines and metanephrines. J Clin Hypertens. 2020;22:1924–1931. 10.1111/jch.14014